Multimodal functional network connectivity: an EEG-fMRI fusion in network space

Area of Science:

• Neuroscience
• Biomedical Engineering
• Data Science

Background:

• Electroencephalography (EEG) and functional magnetic resonance imaging (fMRI) measure brain activity using neuroelectric and hemodynamic signals, respectively.
• Understanding interactions between distributed brain networks is crucial for deciphering complex neurological relationships.

Purpose of the Study:

• To propose a novel method, multimodal functional network connectivity (mFNC), for fusing EEG and fMRI data in a network space.
• To enhance the analysis of brain functional networks by integrating complementary neuroelectric and hemodynamic information.

Main Methods:

• Functional networks (FNs) were extracted from EEG and fMRI data separately using spatial independent component analysis (ICA).
• Interactions among FNs within each modality were analyzed using Granger causality analysis (GCA).
• fMRI FNs were spatially matched to EEG FNs using network-based source imaging (NESOI) for multimodal fusion.

Main Results:

• The proposed mFNC method successfully revealed underlying neural networks for each modality individually and in combination.
• Investigations using both synthetic and real data confirmed the potential of mFNC in identifying network interactions.

Conclusions:

• mFNC provides a powerful approach for integrating EEG and fMRI data, enabling a comprehensive exploration of functional brain networks.
• This method holds promise for deeper investigation into neural activities and metabolic responses during various tasks and neurological states.